Intelligent construction method and system for highway super-large section tunnel

Abstract

The application discloses a kind of highway super large section tunnel intelligent construction method and system, method includes: in the drilling process of highway super large section tunnel, the drilling parameter set of working face is collected;According to the drilling parameter set, determine multiple groups of matching drilling parameters, and determine the drilling parameters of different sections;Based on the drilling parameter set and the drilling parameters of different sections, utilize the pre-trained surrounding rock classification model based on machine learning, determine the overall surrounding rock grade of tunnel and the surrounding rock grade of different sections;According to the overall surrounding rock grade and the surrounding rock grade of different sections, corresponding tunnel support scheme is generated, to realize the intelligent construction of tunnel.Utilize the embodiment of the application, can be combined with artificial intelligence to realize the rapid evaluation of surrounding rock grade, generate corresponding support scheme, improve the safety and efficiency of construction.

Description

TECHNICAL FIELD

[0001] The application belongs to the technical field of tunnel construction, and particularly relates to an intelligent construction method and system for a highway super-large-section tunnel. BACKGROUND

[0002] Tunnel engineering is essentially a geological engineering. In the construction process of a tunnel, various different stratum environments are encountered, and various geological disasters are inevitably encountered in the tunneling process. Tunnel construction is performed underground, and therefore there are many uncertain safety influencing factors. In the tunnel construction process, collapse is one of the most common safety accidents.

[0003] Once a safety accident such as collapse occurs in tunnel engineering, not only is the construction period delayed and the engineering cost greatly increased, but also personal injury or even life-threatening to construction workers and technical personnel is caused. Therefore, support measures are very crucial in the tunnel construction process. The effectiveness of the support measures is naturally higher the more sufficient the support measures are, the higher the safety is, but the construction cost is greatly increased, the construction progress is slowed down, and even the size of the tunnel itself is affected. Therefore, balancing the safety and cost of the support measures is a problem that needs to be faced. Timely and effective support measures usually depend on accurate surrounding rock classification and expert experience. How to determine the support measures by accurate surrounding rock classification in the construction process to reduce the dependence on expert experience is a problem that needs to be solved at present. SUMMARY

[0004] The application aims to provide an intelligent construction method and system for a highway super-large-section tunnel to solve the problems in the prior art, and to realize rapid evaluation of surrounding rock grades by combining artificial intelligence, generate a corresponding support scheme, and improve the safety and efficiency of construction.

[0005] One embodiment of the present application provides an intelligent construction method for a highway super-large-section tunnel, which comprises the following steps:

[0006] Collecting a set of drilling parameters of a tunnel face in the drilling process of the highway super-large-section tunnel;

[0007] Determining a plurality of groups of paired drilling parameters according to the set of drilling parameters, and determining drilling parameters of different sections;

[0008] Determining overall surrounding rock grades of the tunnel and surrounding rock grades of different sections by using a pre-trained surrounding rock classification model based on machine learning based on the set of drilling parameters and the drilling parameters of different sections;

[0009] Generating a corresponding tunnel support scheme according to the overall surrounding rock grades and the surrounding rock grades of different sections to realize intelligent construction of the tunnel.

[0010] Optionally, the drilling parameter set of the tunnel face is collected, including:

[0011] The tunnel face and its surrounding area are meshed and divided into a plurality of sub-zones according to actual geological characteristics and drilling progress, wherein each sub-zone is responsible for collecting parameter data of each drilling parameter in the sub-zone, and the types and quantities of drilling parameters collected by each sub-zone are the same, and the division of the sub-zones considers the uniformity of the rock stratum and the change of the surrounding rock so as to record the drilling characteristics of each sub-zone;

[0012] The parameter data of the drilling parameters collected by each sub-zone are combined to form a drilling parameter set of the final tunnel face.

[0013] Optionally, the drilling parameters of different sections are determined according to the drilling parameter set, including:

[0014] In the sub-zone, a paired parameter generation mechanism is established according to the drilling parameters of adjacent sub-zones, and a sliding window algorithm is used to gradually slide the window and collect the drilling parameters of adjacent sub-zones to generate a paired drilling parameter set;

[0015] The drilling parameters of each paired sub-zone are converted into a feature vector, and the feature vector is mapped to a nonlinear feature space;

[0016] In the nonlinear feature space, a clustering algorithm is applied to classify all feature vectors to obtain feature vectors of each classification, and the drilling parameters corresponding to the feature vectors of each classification are determined as the drilling parameters of each section.

[0017] Optionally, the drilling parameters of different sections are determined according to the drilling parameter set, including:

[0018] A sliding window is created, the size of the sliding window is set by a user, and the sliding window contains a plurality of adjacent sub-zones, so that complete drilling parameter information of the sub-zones can be obtained each time the sliding window is moved;

[0019] In each time step, the drilling parameters of each sub-zone contained in the sliding window are obtained each time the window position of the sliding window is moved by one step;

[0020] The drilling parameters of each sub-zone are arranged in a vector form and standardized, and the standardized drilling parameters of each pair of adjacent sub-zones in the window are judged for similarity;

[0021] If it is judged that the similarity is less than the preset similarity threshold, the pairing relationship of the pair of adjacent sub-areas is recorded, and the drilling parameters of the pair of adjacent sub-areas are recorded in the paired drilling parameter set. The pairing relationship of all adjacent sub-areas is obtained through the sliding window, and a complete paired drilling parameter set is finally formed for analyzing the overall construction state and the surrounding rock characteristics.

[0022] Optionally, based on the drilling parameter set and the drilling parameters of different sections, a pre-trained surrounding rock classification model based on machine learning is used to determine the overall surrounding rock grade of the tunnel and the surrounding rock grades of different sections, including:

[0023] The drilling parameter set is input into a pre-trained first surrounding rock classification model based on machine learning to determine the overall surrounding rock grade of the tunnel, wherein the first surrounding rock classification model is trained based on historical drilling parameter sets and corresponding historical overall surrounding rock grades.

[0024] The drilling parameters of different sections are respectively input into a pre-trained second surrounding rock classification model based on machine learning to determine the surrounding rock grades of different sections of the tunnel, wherein the second surrounding rock classification model is trained based on historical drilling parameters of different sections and corresponding historical surrounding rock grades of different sections.

[0025] The surrounding rock grades of different sections are weighted to obtain a weighted overall surrounding rock grade.

[0026] According to the overall surrounding rock grade determined by the first surrounding rock classification model and the weighted overall surrounding rock grade, a final overall surrounding rock grade of the tunnel is determined.

[0027] Another embodiment of the present application provides a highway super-large section tunnel intelligent construction system, the system comprising:

[0028] A collection module is configured to collect a drilling parameter set of a working face during drilling of a highway super-large section tunnel.

[0029] A first determination module is configured to determine a plurality of paired drilling parameters based on the drilling parameter set and determine drilling parameters of different sections.

[0030] A second determination module is configured to use a pre-trained surrounding rock classification model based on machine learning to determine the overall surrounding rock grade of the tunnel and the surrounding rock grades of different sections based on the drilling parameter set and the drilling parameters of different sections.

[0031] A generation module is configured to generate a corresponding tunnel support scheme based on the overall surrounding rock grade and the surrounding rock grades of different sections to realize intelligent construction of the tunnel.

[0032] Still another embodiment of the present application provides a storage medium having a computer program stored therein, wherein the computer program is configured to execute the method described in any of the above embodiments when run.

[0033] Still another embodiment of the present application provides an electronic device comprising a memory having a computer program stored therein and a processor configured to execute the computer program to perform the method described in any of the above embodiments.

[0034] Compared with the prior art, the intelligent construction method of the highway super-large section tunnel provided by the present application can collect a set of drilling parameters of the tunnel face in the drilling process of the highway super-large section tunnel, determine a plurality of groups of matched drilling parameters according to the set of drilling parameters, and determine drilling parameters of different sections; based on the set of drilling parameters and the drilling parameters of different sections, an overall surrounding rock grade of the tunnel and surrounding rock grades of different sections are determined by using a pre-trained surrounding rock classification model based on machine learning; and a corresponding tunnel support scheme is generated according to the overall surrounding rock grade and the surrounding rock grades of different sections to realize intelligent construction of the tunnel, so that the surrounding rock grade can be quickly evaluated in combination with artificial intelligence, the corresponding support scheme is generated, and the safety and efficiency of construction are improved. BRIEF DESCRIPTION OF DRAWINGS

[0035] Figure 1 A hardware structure block diagram of a computer terminal of the intelligent construction method of the highway super-large section tunnel provided by the embodiment of the present application is provided.

[0036] Figure 2 A flowchart of the intelligent construction method of the highway super-large section tunnel provided by the embodiment of the present application is provided.

[0037] Figure 3 A structure diagram of the intelligent construction system of the highway super-large section tunnel provided by the embodiment of the present application is provided. DETAILED DESCRIPTION

[0038] The embodiments described below with reference to the drawings are exemplary and are only used to explain the present application, and cannot be explained as a limitation of the present application.

[0039] The embodiment of the present application first provides an intelligent construction method of a highway super-large section tunnel, which can be applied to an electronic device such as a computer terminal, specifically, a general computer and the like.

[0040] The following will be described in detail taking the running on the computer terminal as an example. Figure 1 A hardware structure block diagram of a computer terminal of the intelligent construction method of the highway super-large section tunnel provided by the embodiment of the present application is provided. As shown in the figure, Figure 1As shown in the figure, the computer device includes a processor, a memory and a network interface connected through a system bus, wherein the memory can include a non-volatile storage medium and an internal memory.

[0041] The non-volatile storage medium can store an operating system and a computer program. The computer program includes program instructions which, when executed, can cause the processor to perform any one of the intelligent construction methods of the highway super-large section tunnel.

[0042] The processor is used to provide computing and control capabilities to support the operation of the entire computer device.

[0043] The internal memory provides an environment for the running of the computer program in the non-volatile storage medium, which, when executed by the processor, can cause the processor to perform any one of the intelligent construction methods of the highway super-large section tunnel.

[0044] The network interface is used for network communication, such as sending assigned tasks, etc. Those skilled in the art can understand that, Figure 1 The structure shown in the figure is only a block diagram of part of the structure related to the scheme of the present application, and does not constitute a limitation on the computer device to which the scheme of the present application is applied. The specific computer device can include more or fewer components than those shown in the figure, or combine certain components, or have a different component arrangement.

[0045] It should be understood that the processor can be a central processing unit (CPU), and the processor can also be other general-purpose processors, digital signal processors (DSP), application specific integrated circuits (ASIC), field programmable gate arrays (FPGA) or other programmable logic devices, discrete gates or transistor logic devices, discrete hardware components, etc. Among them, the general-purpose processor can be a microprocessor or the processor can also be any conventional processor, etc.

[0046] Referring to Figure 2 The embodiments of the present application provide an intelligent construction method of a highway super-large section tunnel, which can include the following steps:

[0047] S201, collecting a set of drilling parameters of a working face in the drilling process of the highway super-large section tunnel;

[0048] In the construction of highway super-large section tunnels, the tunnel face is the most forward working area during the construction process, directly affected by the external geological environment. By collecting drilling parameter sets in this area, a series of key construction data can be obtained, such as drilling speed, torque, drilling pressure, footage, etc. Real-time monitoring of these parameters provides a basis for subsequent data analysis and surrounding rock classification, reflecting the characteristics and trends of the current surrounding rock. Therefore, collecting drilling parameter sets at the tunnel face is one of the core steps in intelligent construction, ensuring dynamic control of the construction state.

[0049] Collecting drilling parameter sets at the tunnel face not only allows real-time tracking of various changes during the construction process, but also helps engineers identify potential problems in a timely manner and take appropriate measures to ensure construction safety. This process provides important basis for subsequent surrounding rock level evaluation and support scheme generation, thereby optimizing the construction scheme, reducing construction costs, and improving engineering efficiency. Therefore, a deep understanding and use of drilling parameter sets is of great significance for the intelligent development of tunnel construction.

[0050] Specifically, collecting drilling parameter sets at the tunnel face can divide the tunnel face and its surrounding area into a grid, and divide it into multiple sub-areas according to the actual geological characteristics and drilling progress. Each sub-area is responsible for collecting parameter data for each type of drilling parameter in that sub-area, and each sub-area collects the same types and quantities of drilling parameters. The division of the sub-area takes into account the uniformity of the rock stratum and the change of the surrounding rock, in order to record the drilling characteristics of each sub-area.

[0051] In the construction of highway super-large section tunnels, the grid division of the tunnel face and its surrounding area is the basic step of parameter collection. By dividing the excavation area into multiple sub-areas according to the actual geological characteristics and drilling progress, each sub-area can systematically collect various drilling parameters. This division is not arbitrary, but takes into account the uniformity of the rock stratum, the geological characteristics, and the change of the surrounding rock, to ensure that representative drilling characteristics can be recorded in different sub-areas.

[0052] The introduction of grid division can effectively improve the understanding of the geological conditions of the tunnel face, ensuring the comprehensiveness and consistency of data in different areas. This method not only helps to analyze the characteristics of the surrounding rock in each sub-area, but also provides important data support for developing more scientific support schemes. Through standardized regional division, the construction team can monitor specific parameters in real time for each sub-area, thereby achieving more efficient construction management.

[0053] When implementing the grid division, first, on-site geological exploration is needed to collect geological data of the working face and its surrounding area, including rock type, strength, fracture distribution, and underground water level, etc. Based on these data, engineers can use GIS (Geographic Information System) software to visualize the working face and develop a reasonable division plan. During the division process, the working face and its surrounding area are usually divided into several small blocks (subareas) of equal or similar size, such as 2-meter diameter circular subareas, to ensure that each subarea is representative in terms of geological characteristics.

[0054] After the division is completed, each subarea will be equipped with the same drilling parameter acquisition equipment, such as pressure sensors, displacement sensors, and temperature sensors, etc., to ensure that drilling parameter data of each subarea can be collected in real time and accurately during the drilling process. It is worth noting that the division of subareas should be timely tracked and adjusted to adapt to the changes in construction progress and surrounding rock conditions. For example, when the surrounding rock conditions of a subarea change significantly, the number of subareas can be increased, or the division of subareas can be adjusted to facilitate more accurate data processing.

[0055] The parameter data of drilling parameters collected from each subarea form the final working face drilling parameter set.

[0056] After dividing and independently obtaining drilling parameters from multiple subareas, the next step is to integrate these parameters to form a complete working face drilling parameter set. This process is crucial and involves aggregating, organizing, and forming a unified format for data from each subarea to facilitate subsequent analysis and processing.

[0057] By integrating the parameters of each subarea, a comprehensive construction status view can be formed, providing a systematic data basis for the evaluation of surrounding rock characteristics. This method ensures that the parameters of each subarea can be compared and analyzed with each other, reducing inconsistencies and improving data reliability. Ultimately, this integration work provides a solid foundation for the training of machine learning models and the evaluation of surrounding rock grades.

[0058] When integrating drilling parameters, first, the data collected from each subarea needs to be subjected to preliminary quality checks to ensure data accuracy and completeness. This process usually includes removing noise and outliers, and supplementing missing data. After data cleaning, use spreadsheet or database tools to establish a central database to collect and store parameter data of all subareas.

[0059] Next, the drilling parameters of all subzones are input into the database in a unified format. For example, they can be classified by parameter type (such as drilling speed, pressure, footage, etc.), each parameter has its standard unit, and the corresponding timestamp and subzone number are recorded. Through data management software, it can ensure that the data of various parameters has continuity in time, and can effectively track the drilling progress of each subzone.

[0060] After the data integration is completed, engineers will use statistical analysis and visualization tools to analyze the aggregated data to extract key features and trends. These analysis results not only help the construction team to understand the changes in geological conditions in a timely manner, but also provide data support for subsequent surrounding rock assessment and support schemes. For example, by normalizing the parameters of different subzones, it can identify which subzones have dangerous surrounding rock conditions, so that appropriate reinforcement measures can be taken in advance.

[0061] S202, determine a plurality of paired drilling parameters according to the drilling parameter set, and determine drilling parameters of different sections;

[0062] During tunnel construction, changes in drilling parameters can reflect important information such as changes in surrounding geological conditions, working status of construction equipment, and construction progress. Therefore, by effectively pairing the drilling parameters of the working face and different sections, it is helpful to identify and analyze the construction performance under different construction stages and different geological conditions. By establishing multiple sets of paired drilling parameters, construction personnel can better understand and master the dynamic changes in the construction process, so as to make timely adjustments and optimizations. This pairing is not limited to parameters within the same time period, but also considers parameter changes in adjacent time periods to ensure that the formed parameter set covers more comprehensive construction state information.

[0063] Through the paired analysis of drilling parameters of different sections and subzones, a depth analysis of the geological surrounding rock state can be achieved. This process enables the construction team to more accurately assess the characteristics of surrounding rock when faced with complex geological conditions, thereby developing more reasonable and scientific construction schemes. At the same time, this can also improve the safety and efficiency of overall tunnel construction, reduce construction risks, and ensure smooth project progress. In addition, good data pairing not only provides support for current construction, but also lays the foundation for future engineering optimization and technical improvement.

[0064] Specifically, within the subzone, a paired parameter generation mechanism can be established according to the drilling parameters of adjacent subzones, and a sliding window algorithm can be used to gradually slide the window and collect drilling parameters of adjacent subzones to generate a set of paired drilling parameters;

[0065] The core of this step is to use the sliding window algorithm to establish connections between various sub-zones and collect drilling parameters of adjacent sub-zones. This method can effectively capture the continuity and variability in the drilling process, thereby generating multiple pairs of correlated parameters. By adopting a sliding window, it can be ensured that the data collected each time is complete information within a time period, maximizing the representativeness and accuracy of the collected data.

[0066] Pairing analysis of drilling parameters in adjacent sub-zones not only makes the relationship between each sub-zone more closely, but also provides a basis for comprehensive evaluation of the construction state of the entire tunnel. By analyzing these paired parameters, the construction team can clearly identify the changes in the surrounding rock characteristics in different areas, and then develop appropriate construction strategies and support measures. This mechanism can effectively reduce the blindness in the construction process, improve the response speed, and ensure the efficiency and safety of tunnel construction.

[0067] Specifically, a sliding window can be created, the size of which is set by the user, and the sliding window contains multiple adjacent sub-zones, so that each sliding can obtain complete drilling parameter information of the sub-zones;

[0068] In this step, the user sets the size of the sliding window according to actual needs, and the window contains multiple adjacent sub-zones, so as to efficiently collect relevant drilling parameters during construction. The design of the sliding window allows complete information of adjacent sub-zones to be obtained at each sliding, ensuring the continuity and completeness of the data, which is helpful for subsequent analysis and decision-making.

[0069] By flexibly setting the size of the sliding window, the system can adapt to different construction environments and needs, improving the pertinence and flexibility of data collection. At the same time, the design of the window containing multiple sub-zones can ensure that the data obtained is more comprehensive, thereby improving the accuracy and effectiveness of subsequent analysis. This provides important data support for scheduling and decision-making during construction.

[0070] First, the user can set the size of the sliding window according to the actual drilling operation needs, for example, set the window to cover 5 adjacent sub-zones. Assuming that in a certain tunnel construction site, the sub-zones are A, B, C, D and E in turn. When creating the window, the system will initialize to cover A, B, C, D and E. In this window position, the system will collect the drilling parameters of the 5 sub-zones in real time when the construction personnel perform drilling operations. If the window size is set to 5, when the window starts to slide, the window position will be moved to B, C, D, E and F, so that the latest drilling parameters of each sub-zone can be obtained at each sliding. Thus, the sliding window can continuously update the sub-zone information it contains, thereby ensuring that the collected data is complete and suitable for subsequent analysis.

[0071] At each time step, the window position of the sliding window is moved one grid, and the drilling parameters of each sub-area contained in the current window are obtained;

[0072] In this step, the system automatically moves the window position by one grid according to the set time step, and obtains the drilling parameters of each sub-area in the current window. This ensures that the data is updated in real time at each sliding, accurately reflecting the construction status and parameter changes in different time periods.

[0073] By sliding the window by time step, dynamic monitoring of drilling parameters can be achieved. This process can capture changes in construction in a timely manner, helping managers quickly identify problems and make adjustments. At the same time, this dynamic data acquisition method provides more rich time series data for subsequent analysis, supporting real-time monitoring and evaluation.

[0074] In actual operation, assume that the set time step is 1 hour, and the system will move the sliding window one grid to the right every 1 hour. For example, if the current window covers A, B, C, D, and E, after 1 hour, the window will move to B, C, D, E, and F. At this time, the system begins to collect the drilling parameters of each sub-area in the current window, including drilling rate, pressure, torque, etc. During the collection process, the system will automatically record the parameters of each sub-area at that time point, providing a basis for subsequent data processing and analysis. This method ensures that all parameters are compared based on the same time step, which helps subsequent similarity judgment and other analysis.

[0075] Arrange the drilling parameters of each sub-area into a vector form and perform standardization, and perform similarity judgment on the standardized drilling parameters of each pair of adjacent sub-areas in the window;

[0076] In this step, the system arranges the drilling parameters of each sub-area into a vector form and performs standardization. The purpose of this step is to convert the data of different sub-areas into comparable standards for subsequent similarity judgment, making data analysis more scientific and accurate.

[0077] By arranging the drilling parameters into a vector and performing standardization, the unfair comparison caused by different parameter magnitudes can be eliminated. This step lays the foundation for similarity judgment between adjacent sub-areas, ensuring that more real and effective analysis results can be obtained in subsequent comparisons, which helps better understand the problems and reasons in construction.

[0078] In implementation, the system first organizes the drilling parameters (such as drilling rate, pressure, etc.) of each sub-area into vector form. For example, assuming that the drilling rates of sub-areas A, B, and C in the current window are 6 m / h, 4 m / h, and 5 m / h respectively, the organized vectors will be A = [6, x, y], B = [4, x, y], and C = [5, x, y] (where x and y are other parameters). Next, the system will perform standardization processing on these vectors, for example, using the Z-score method, so that the mean of each parameter is 0 and the standard deviation is 1. After standardization, the resulting vectors will be A', B', and C'. Next, the system will evaluate the parameter similarity of each pair of adjacent sub-areas by calculating similarity indicators such as cosine similarity or Euclidean distance. In this way, it can effectively identify which sub-areas have similar parameters, providing a basis for subsequent pairing relationship records.

[0079] If it is determined that the similarity is less than the preset similarity threshold, the pairing relationship of the pair of adjacent sub-areas is recorded, and the drilling parameters of the pair of adjacent sub-areas are recorded in the paired drilling parameter set. The pairing relationship of all adjacent sub-areas is continuously obtained through the sliding window, and a complete set of paired drilling parameters is finally formed to analyze the overall construction state and surrounding rock characteristics.

[0080] In this step, the system records the sub-areas and their drilling parameters in the paired drilling parameter set if it finds that the similarity of a pair of adjacent sub-areas is less than the preset similarity threshold. Such pairing relationship helps to further analyze different sub-areas and identify potential problems and their impact on construction.

[0081] By recording the pairing relationship with similarity less than the preset threshold, significant differences between different sub-areas can be found. This process helps the construction management team to analyze the surrounding rock characteristics and overall construction state in depth, so as to timely adjust the construction plan, optimize resource allocation, and ensure the smooth progress and safety of the project.

[0082] In implementation, the system sets a similarity threshold, for example, 0.7. After similarity judgment is completed, assuming that the similarity of A and B is 0.6, the pair of sub-areas meets the recording condition. The system will automatically record the identification of adjacent sub-areas A and B, their respective standardized parameter vectors, and the similarity score in the paired drilling parameter set. This set can be in the form of a database or a data table, recording the pairing information of each pair of adjacent sub-areas, including their drilling rate, pressure value, etc. As the sliding window moves continuously, the system will continuously update and expand this set. When all the pairing relationship records are completed, engineers can analyze the set to compare the construction efficiency and surrounding rock characteristics of different sub-areas, and thus develop targeted treatment schemes to ultimately improve construction safety and efficiency.

[0083] Converting the drilling parameters of each paired sub-zone into a feature vector, and mapping the feature vector to a non-linear feature space;

[0084] The main task of this step is to convert drilling parameters into feature vectors for further analysis and processing. The process of mapping to a non-linear feature space can help reveal more complex and potential relationships between parameters in high-dimensional space.

[0085] Converting drilling parameters into feature vectors and mapping them to a non-linear feature space not only provides a basis for subsequent clustering analysis, but also enhances the model's ability to learn complex non-linear relationships. This process can improve the accuracy of subsequent machine learning algorithms, provide more rich and accurate feature data for surrounding rock classification models, and thus promote the evaluation of surrounding rock grades and the optimization of support schemes.

[0086] After the data is sorted, first create a feature vector for each paired sub-zone. For example, drilling speed, drilling pressure, displacement and other parameters can be used as features to form a multi-dimensional feature vector in each adjacent sub-zone. Then, through a specific mapping function (such as a polynomial kernel function or a radial basis function), these feature vectors are mapped to a non-linear feature space. During the mapping process, the system constructs new feature combinations based on the relationships between the original features to more accurately capture the non-linear relationships between the parameters.

[0087] In the non-linear feature space, apply a clustering algorithm to classify all feature vectors, and obtain the feature vectors of each classification. The drilling parameters corresponding to the feature vectors of each class are determined as the drilling parameters of each section.

[0088] By using a clustering algorithm to classify feature vectors, the goal is to identify and group drilling parameters with similar features. Successful implementation of this step can clearly divide the drilling parameters of different sections, providing an accurate basis for subsequent evaluation of surrounding rock grades.

[0089] By clustering similar feature vectors into a class, it helps to identify the commonalities and differences in the drilling process of different sections. This analysis result can guide the construction team to develop more targeted construction plans and support measures for different areas. Ultimately, the feature vectors of each classification will be directly used to evaluate and optimize the characteristics of the surrounding rock of the tunnel, which is of great significance to improve overall construction safety and efficiency.

[0090] During the clustering analysis phase, common clustering algorithms such as K-means clustering, hierarchical clustering or DBSCAN can be selected. According to the requirements, set the parameters of clustering, such as the number of clusters or distance measurement methods. Then, input the feature vectors converted previously into the selected clustering algorithm.

[0091] The algorithm will group them based on the similarity between the feature vectors. Each clustering result will form an independent category, and each category will contain its corresponding drilling parameters. By analyzing the features of each cluster, the cross-section drilling parameters represented by the feature vectors of the category can be determined. This process needs to be verified and adjusted to ensure the effectiveness and accuracy of the classification results, and finally form a clear set of cross-section drilling parameters to support subsequent surrounding rock assessment and support design.

[0092] S203, based on the drilling parameter set and the drilling parameters of different sections, using a pre-trained machine learning-based surrounding rock classification model, determining the overall surrounding rock grade of the tunnel and the surrounding rock grade of different sections;

[0093] In this method, the drilling parameter set and the drilling parameters of different sections collected during the drilling process of the highway super-large cross-section tunnel are used to determine the overall surrounding rock grade of the tunnel using a pre-trained machine learning-based surrounding rock classification model. This process involves inputting two types of parameters into the trained model, which has learned a large amount of historical data and can identify and predict the properties and behavior of surrounding rock. In the model operation, different input features such as drilling rate, pressure and torque are used for feature extraction and prediction analysis, and finally the overall and different section surrounding rock grades of the tunnel are output, providing a basis for subsequent support scheme formulation.

[0094] The key to this step is to automate the complex surrounding rock classification process through the application of machine learning models, improving the scientific nature and efficiency of decision-making in the construction process. Through the analysis of multiple groups of drilling parameters, the properties and levels of surrounding rock can be more accurately judged, thereby providing important reference for the construction process and ensuring the safety and effectiveness of the construction. In addition, through this data-driven approach, the surrounding rock classification model can be continuously optimized to provide more reliable data support and guidance for future similar construction.

[0095] Specifically, the drilling parameter set can be input into a pre-trained first machine learning-based surrounding rock classification model to determine the overall surrounding rock grade of the tunnel, wherein the first surrounding rock classification model is trained based on historical drilling parameter sets and corresponding historical overall surrounding rock grades;

[0096] In this step, the collected drilling parameter set will be input as input data into the first surrounding rock classification model that has been pre-trained. This model is constructed based on historical drilling parameters and corresponding surrounding rock grades, and through comprehensive learning of historical data, the model can identify the complex relationship between drilling parameters and surrounding rock grades, and can use new data to make inferences to obtain the classification of the overall surrounding rock grade of the tunnel.

[0097] By inputting the data into a specialized machine learning model, the accuracy and efficiency of surrounding rock classification can be greatly improved. Compared to traditional experience-based evaluation methods, machine learning models can quickly process large amounts of data and extract key features that affect the properties of surrounding rock, providing more scientific and reasonable decision-making basis for construction management. In addition, this method helps to realize real-time monitoring and dynamic adjustment during tunnel construction, improving engineering safety.

[0098] In this process, first of all, the collected drilling parameter set needs to be sorted and pre-processed, including drilling speed, torque, pressure, etc. Then, these data are formatted into the input format required by the model, and are sent to the first surrounding rock classification model. Assuming that in the past construction, historical data of multiple tunnel projects are collected, including the corresponding surrounding rock grades of each project. The model can make judgments based on the input drilling parameter characteristics after learning and training on these historical data. For example, if the drilling speed represented by the input data is high and the pressure fluctuation is small, the model will infer that the surrounding rock grade of this area may be high based on previous learning. The processed output data will give a classification of the overall surrounding rock grade, such as 1st (excellent), 2nd (good), 3rd (medium), 4th (poor) grade, for engineers to reference for making subsequent construction plans and support measures.

[0099] The drilling parameters of different sections are input into the pre-trained second surrounding rock classification model based on machine learning to determine the surrounding rock grades of different sections of the tunnel, wherein the second surrounding rock classification model is trained based on historical drilling parameters of different sections and corresponding historical surrounding rock grades of different sections;

[0100] In this step, the drilling parameters of different sections are input into the trained second surrounding rock classification model to evaluate the surrounding rock grade of each section. The second surrounding rock classification model is specially trained based on historical data of different sections, and can analyze the surrounding rock characteristics and changes under specific conditions for each section. Through this process, the obtained surrounding rock grade provides specific technical support for processing different sections.

[0101] By processing the drilling parameters of different sections independently, the properties of the surrounding rock of each section and its impact on construction can be more accurately understood. This evaluation helps to develop more reasonable support schemes and construction strategies, thereby improving the safety and effectiveness of the entire construction process. In addition, using machine learning models for evaluation is more efficient and accurate than traditional methods, and can timely feedback the changes in the surrounding rock state of each section, facilitating dynamic adjustment by the construction team.

[0102] In the step implementation, the construction team first needs to classify and organize the drilling parameters of different sections, including drilling speed, pressure, and torque parameters collected for each section. Then, these data will be formatted and input into the trained second surrounding rock classification model. Assuming that a certain section exhibits high drilling pressure and low drilling speed during construction, it may be determined to have a low surrounding rock grade based on the model's historical learning, requiring enhanced support. After evaluating each section, the model will provide the surrounding rock grade for that section based on the input parameters, allowing engineers to develop specific optimized construction plans for different sections to ensure smooth construction.

[0103] The surrounding rock grades of the different sections are weighted to obtain a weighted overall surrounding rock grade.

[0104] In this step, the surrounding rock grades of each section obtained through the second surrounding rock classification model are weighted to generate a weighted overall surrounding rock grade. This process takes into account the different importance of each section and provides more accurate overall tunnel surrounding rock conditions through weighted evaluation, providing more effective data support for the development of construction plans.

[0105] The weighted overall surrounding rock grade is more representative and can more accurately reflect the overall tunnel surrounding rock characteristics. This method can help the construction team optimize the construction process, reasonably allocate resources, and ensure construction safety and efficiency under complex geological conditions. Meanwhile, using weighted evaluation makes the model more flexible and adaptable in practical applications.

[0106] In the implementation of weighted processing, the construction team needs to first determine the weight of each section, which is usually based on the geological characteristics of the section, historical construction conditions, or importance evaluation of the entire tunnel. Assuming that a certain section is assigned a higher weight due to its complex geology, while other relatively simple sections are assigned lower weights. Next, multiply the surrounding rock grade of each section by its weight to obtain the weighted value. Then, sum up the weighted values of all sections to obtain a weighted overall surrounding rock grade. Assuming that Section 1 has a grade of 1 and a weight of 0.4, and Section 2 has a grade of 2 and a weight of 0.6, the overall grade is calculated as Overall Grade = 1 × 0.4 + 2 × 0.6 = 1.6, which can be understood as good to slightly better (i.e., slightly better). The result will provide a scientific basis for subsequent tunnel support plans.

[0107] Based on the overall surrounding rock grade determined by the first surrounding rock classification model and the weighted overall surrounding rock grade, the final overall surrounding rock grade of the tunnel is determined.

[0108] In the final step, the overall surrounding rock grade obtained from the first surrounding rock classification model is combined with the overall surrounding rock grade obtained after weighting processing, and the final overall surrounding rock grade of the tunnel is comprehensively evaluated and determined. This process aims to integrate the results obtained from different models to form a more accurate and comprehensive evaluation, providing direct basis for the development of construction strategies.

[0109] By combining the results of different models, the bias that may exist in a single model can be eliminated in the evaluation, improving the accuracy and reliability of the overall evaluation. This integration step ensures that in complex tunnel construction environment, the construction team can develop targeted and highly operational construction schemes, further improving project safety and construction efficiency.

[0110] In this implementation process, first, the overall surrounding rock grade obtained from the first surrounding rock classification model is compared and analyzed with the weighted overall surrounding rock grade. The construction team will use certain rules or algorithms, such as simple weighted average or maximum value method, to determine the final overall surrounding rock grade. Assuming that the first model obtains an overall surrounding rock grade of "good", and the weighted model obtains an overall surrounding rock grade of "medium", then a compromise evaluation may be made, determining "good-medium". This comprehensive evaluation result will provide direct reference basis for subsequent construction decision-making, ensuring that the construction process is more scientific and reasonable.

[0111] S204, according to the overall surrounding rock grade and the surrounding rock grade of different sections, a corresponding tunnel support scheme is generated to realize intelligent construction of the tunnel.

[0112] According to the overall surrounding rock grade and the surrounding rock grade of different sections, this method aims to generate corresponding tunnel support schemes to realize intelligent construction of the tunnel. Specifically, by analyzing the determined surrounding rock grade, the construction team can identify the surrounding rock characteristics of different sections and the overall tunnel, thereby providing basis for subsequent support design. The support scheme will be designed according to the surrounding rock conditions of each section, ensuring that the designed support structure can meet safety requirements and improve construction efficiency, ultimately forming a systematic and intelligent tunnel construction process.

[0113] The generated tunnel support scheme has important practical significance. First, by using machine learning models to evaluate the surrounding rock grade, the construction party can make more scientific support decisions based on data, thereby improving construction safety and reducing potential risks and accident rates. Second, the tailor-made support scheme can optimize resource allocation, reduce unnecessary material waste, and improve construction progress, ultimately reducing the overall cost of the project. At the same time, this intelligent technology-based construction method also provides a reference standard for future tunnel construction, promoting the progress of industry technology.

[0114] In the implementation process, first, the construction team will systematically analyze the overall surrounding rock level and the surrounding rock level of different sections based on the data obtained from the previous steps. Assuming that the overall surrounding rock assessment is "good", while a specific section is rated as "poor", the construction team needs to develop a corresponding support scheme for the "poor" section. The specific operation steps can be as follows:

[0115] 1. Data integration and evaluation: Collect the overall and individual section surrounding rock levels and integrate them into a comprehensive evaluation report. This includes specific parameters of each section, historical surrounding rock conditions and current evaluation results. In this report, it is clear that the section and its surrounding rock characteristics that need special attention, such as cracks, loose soil or hydrological conditions that may exist in the section.

[0116] 2. Preliminary design of support scheme: Determine the support type according to the evaluation report. If a section is rated as "poor", the team may choose steel support and sprayed concrete, while for "good" sections, simple pre-stressed anchor support can be used. The team will calculate the size, quantity and materials of the support structure in detail to ensure the effectiveness and stability of the support scheme.

[0117] 3. Dynamic adjustment and feedback mechanism: After the preliminary design of the support scheme is completed, the construction team also needs to establish a dynamic feedback mechanism. During the tunnel construction process, real-time monitoring of drilling parameters and surrounding rock behavior is required, and if changes in the surrounding rock conditions of a section are found, such as increased pressure or slowed drilling speed, immediate feedback to the support design is required, and the support scheme may need to be adjusted. For example, if the monitoring results show that the surrounding rock strength of a section has decreased, the construction personnel may need to strengthen the support, increase the support points or use higher strength materials.

[0118] 4. Comprehensive technology and material selection: When developing the support scheme, consider using new materials and intelligent sensor technology to improve the safety and durability of the overall structure. For example, Internet of Things sensors can be embedded to monitor the surrounding rock conditions in real time and feed data back to the central control system to facilitate the construction team to make necessary adjustments during the process. At the same time, advanced simulation technology can be applied to simulate and test the support scheme to evaluate its feasibility and effectiveness.

[0119] Through the above steps, a comprehensive tunnel support scheme is formed, which not only ensures the safety and efficiency of construction, but also improves the overall construction management level through intelligent means, better coping with complex geological and environmental conditions. This strategy also provides experience and foundation for future intelligent tunnel construction.

[0120] It can be seen that the drilling parameter set of the tunnel face is collected in the drilling process of the highway super large section tunnel; a plurality of groups of paired drilling parameters are determined according to the drilling parameter set, and drilling parameters of different sections are determined; based on the drilling parameter set and the drilling parameters of different sections, the overall surrounding rock grade of the tunnel and the surrounding rock grades of different sections are determined by using a pre-trained surrounding rock classification model based on machine learning; and a corresponding tunnel support scheme is generated according to the overall surrounding rock grade and the surrounding rock grades of different sections, so as to realize intelligent construction of the tunnel, thereby being capable of realizing rapid evaluation of the surrounding rock grade in combination with artificial intelligence, generating a corresponding support scheme, and improving safety and efficiency of construction.

[0121] Another embodiment of the present application provides a highway super large section tunnel intelligent construction system, referring to Figure 3 , which can include:

[0122] The collection module 301 is configured to collect a drilling parameter set of a tunnel face in the drilling process of a highway super large section tunnel.

[0123] The first determination module 302 is configured to determine a plurality of groups of paired drilling parameters according to the drilling parameter set, and determine drilling parameters of different sections.

[0124] The second determination module 303 is configured to determine the overall surrounding rock grade of the tunnel and the surrounding rock grades of different sections by using a pre-trained surrounding rock classification model based on machine learning based on the drilling parameter set and the drilling parameters of different sections.

[0125] The generation module 304 is configured to generate a corresponding tunnel support scheme according to the overall surrounding rock grade and the surrounding rock grades of different sections, so as to realize intelligent construction of the tunnel.

[0126] It can be seen that the drilling parameter set of the tunnel face is collected in the drilling process of the highway super large section tunnel; a plurality of groups of paired drilling parameters are determined according to the drilling parameter set, and drilling parameters of different sections are determined; based on the drilling parameter set and the drilling parameters of different sections, the overall surrounding rock grade of the tunnel and the surrounding rock grades of different sections are determined by using a pre-trained surrounding rock classification model based on machine learning; and a corresponding tunnel support scheme is generated according to the overall surrounding rock grade and the surrounding rock grades of different sections, so as to realize intelligent construction of the tunnel, thereby being capable of realizing rapid evaluation of the surrounding rock grade in combination with artificial intelligence, generating a corresponding support scheme, and improving safety and efficiency of construction.

[0127] The embodiment of the present application also provides a storage medium, and the storage medium stores a computer program, wherein the computer program is set to execute the steps in any of the method embodiments when running.

[0128] Specifically, in the present embodiment, the storage medium can be configured to store a computer program for executing the following steps:

[0129] S201, collecting a drilling parameter set of a working face in the drilling process of the highway super-large cross-section tunnel;

[0130] S202, determining a plurality of groups of paired drilling parameters according to the drilling parameter set, and determining drilling parameters of different sections;

[0131] S203, determining an overall surrounding rock grade of the tunnel and surrounding rock grades of different sections based on the drilling parameter set and the drilling parameters of different sections by using a pre-trained machine learning-based surrounding rock grading model;

[0132] S204, generating a corresponding tunnel support scheme according to the overall surrounding rock grade and the surrounding rock grades of different sections to realize intelligent construction of the tunnel.

[0133] It can be seen that in the drilling process of the highway super-large cross-section tunnel, the drilling parameter set of the working face is collected, a plurality of groups of paired drilling parameters are determined according to the drilling parameter set, and drilling parameters of different sections are determined; based on the drilling parameter set and the drilling parameters of different sections, an overall surrounding rock grade of the tunnel and surrounding rock grades of different sections are determined by using a pre-trained machine learning-based surrounding rock grading model; and a corresponding tunnel support scheme is generated according to the overall surrounding rock grade and the surrounding rock grades of different sections to realize intelligent construction of the tunnel, so that the rapid evaluation of the surrounding rock grade can be realized in combination with artificial intelligence, the corresponding support scheme is generated, and the safety and efficiency of construction are improved.

[0134] The embodiment of the present application also provides an electronic device including a memory and a processor, the memory stores a computer program, and the processor is configured to run the computer program to execute the steps in any one of the method embodiments.

[0135] Specifically, the electronic device can further include a transmission device and an input-output device, wherein the transmission device is connected with the processor, and the input-output device is connected with the processor.

[0136] Specifically, in the present embodiment, the processor can be configured to execute the following steps by the computer program:

[0137] S201, collecting a drilling parameter set of a working face in the drilling process of the highway super-large cross-section tunnel;

[0138] S202, determining a plurality of groups of paired drilling parameters according to the drilling parameter set, and determining drilling parameters of different sections;

[0139] S203, determining the overall surrounding rock grade of the tunnel and the surrounding rock grades of different sections by using a pre-trained surrounding rock grading model based on machine learning based on the drilling parameter set and the drilling parameters of different sections;

[0140] S204, generating a corresponding tunnel support scheme according to the overall surrounding rock grade and the surrounding rock grades of different sections to realize intelligent construction of the tunnel.

[0141] It can be seen that the drilling parameter set of the working face is collected in the drilling process of the highway super-large section tunnel; a plurality of groups of matching drilling parameters are determined according to the drilling parameter set, and drilling parameters of different sections are determined; the overall surrounding rock grade of the tunnel and the surrounding rock grades of different sections are determined by using a pre-trained surrounding rock grading model based on machine learning based on the drilling parameter set and the drilling parameters of different sections; and a corresponding tunnel support scheme is generated according to the overall surrounding rock grade and the surrounding rock grades of different sections to realize intelligent construction of the tunnel, so that the rapid evaluation of the surrounding rock grade can be realized by combining artificial intelligence, the corresponding support scheme is generated, and the safety and efficiency of construction are improved.

[0142] The above describes the structure, features and effects of the present application in detail according to the embodiments shown in the drawings. The above description is only the preferred embodiments of the present application, but the present application is not limited by the drawings. Any changes or modifications made in accordance with the concept of the present application, or equivalent embodiments with equivalent changes, are still within the scope of the present application.

Claims

1. A highway super-section tunnel intelligent construction method, characterized in that, The method comprises: Collecting a drilling parameter set of a tunnel face during drilling of a highway super-large section tunnel; wherein the tunnel face and its surrounding area are divided into a plurality of sub-areas by grid division according to actual geological characteristics and drilling progress, wherein each sub-area is responsible for collecting parameter data of each drilling parameter in the sub-area, and the types and quantities of drilling parameters collected by each sub-area are the same, and the division of the sub-areas takes into account the uniformity of the rock stratum and the change of the surrounding rock so as to record the drilling characteristics of each sub-area; and the parameter data of the drilling parameters collected by each sub-area are combined to form a final drilling parameter set of the tunnel face; According to the drilling parameter set, a plurality of paired drilling parameters are determined, and drilling parameters of different sections are determined; wherein, in the sub-area, a paired parameter generation mechanism is established according to the drilling parameters of adjacent sub-areas, and a sliding window algorithm is used to gradually slide the window and collect the drilling parameters of adjacent sub-areas to generate a paired drilling parameter set, which includes: creating a sliding window, the size of the sliding window is set by the user, and the sliding window contains a plurality of adjacent sub-areas, so that each sliding window can obtain complete drilling parameter information of the sub-area; in each time step, the window position of the sliding window is moved by one grid, and the drilling parameters of each sub-area contained in the current window are obtained; the drilling parameters of each sub-area are arranged in a vector form and standardized, and the standardized drilling parameters of each pair of adjacent sub-areas in the window are judged for similarity; if the similarity is less than a predetermined similarity threshold, the pairing relationship of the pair of adjacent sub-areas is recorded, and the drilling parameters of the pair of adjacent sub-areas are recorded in the paired drilling parameter set; the pairing relationship of all adjacent sub-areas is obtained by continuously sliding the window, and a complete paired drilling parameter set is finally formed to analyze the overall construction state and the surrounding rock characteristics; the drilling parameters of each paired sub-area are converted into a feature vector, and the feature vector is mapped to a nonlinear feature space; in the nonlinear feature space, a clustering algorithm is applied to classify all feature vectors to obtain feature vectors of each classification, and the drilling parameters corresponding to the feature vectors of each classification are determined as the drilling parameters of each section; Based on the drilling parameter set and the drilling parameters of different sections, a pre-trained machine learning-based surrounding rock classification model is used to determine the overall surrounding rock grade of the tunnel and the surrounding rock grade of different sections; According to the overall surrounding rock grade and the surrounding rock grade of different sections, a corresponding tunnel support scheme is generated to realize intelligent construction of the tunnel.

2. The method of claim 1, wherein, Based on the drilling parameter set and the drilling parameters of different sections, a pre-trained machine learning-based surrounding rock classification model is used to determine the overall surrounding rock grade of the tunnel and the surrounding rock grade of different sections, which comprises: The drilling parameter set is input into a pre-trained first machine learning-based surrounding rock classification model to determine the overall surrounding rock grade of the tunnel, wherein the first surrounding rock classification model is trained based on historical drilling parameter sets and corresponding historical overall surrounding rock grades. The drilling parameters of the different sections are respectively input into a second surrounding rock classification model based on machine learning which is pre-trained to determine the surrounding rock grades of the different sections of the tunnel, wherein the second surrounding rock classification model is trained based on historical drilling parameters of different sections and historical surrounding rock grades corresponding to the different sections; The surrounding rock grades of the different sections are weighted to obtain a weighted overall surrounding rock grade; The final overall surrounding rock grade of the tunnel is determined according to the overall surrounding rock grade determined by the first surrounding rock classification model and the weighted overall surrounding rock grade.

3. A highway super-section tunnel intelligent construction system applied to the highway super-section tunnel intelligent construction method of claim 1 or 2, characterized in that, The system comprises: A collection module configured to collect a set of drilling parameters of a tunnel face during drilling of a highway super-large cross-section tunnel; A first determination module configured to determine a plurality of pairs of drilling parameters according to the set of drilling parameters and determine drilling parameters of different sections; A second determination module configured to determine an overall surrounding rock grade of the tunnel and surrounding rock grades of different sections by using a pre-trained surrounding rock classification model based on machine learning based on the set of drilling parameters and the drilling parameters of different sections; A generation module configured to generate a corresponding tunnel support scheme according to the overall surrounding rock grade and the surrounding rock grades of different sections to realize intelligent construction of the tunnel.

4. The system of claim 3, wherein, The collection module is specifically configured to: divide the tunnel face and its surrounding area into a plurality of sub-areas according to actual geological characteristics and drilling progress, wherein each sub-area is responsible for collecting parameter data of each drilling parameter in the sub-area, and the types and quantities of drilling parameters collected by each sub-area are the same, and the division of the sub-areas takes into account the uniformity of the rock stratum and the change of the surrounding rock to record the drilling characteristics of each sub-area; combine the parameter data of the drilling parameters collected by each sub-area to form a set of drilling parameters of the final tunnel face.

5. The system of claim 4, wherein, The first determination module is specifically configured to: establish a paired parameter generation mechanism according to the drilling parameters of adjacent sub-areas in the sub-area, and gradually slide the window and collect the drilling parameters of adjacent sub-areas by using a sliding window algorithm to generate a set of paired drilling parameters; convert the drilling parameters of each paired sub-area into a feature vector, and map the feature vector to a nonlinear feature space; apply a clustering algorithm to classify all feature vectors in the nonlinear feature space to obtain feature vectors of each classification, and determine the drilling parameters corresponding to the feature vectors of each classification as the drilling parameters of each section.

6. A storage medium, characterized by The storage medium stores a computer program, wherein the computer program is configured to execute the method of any one of claims 1-2 when running.

7. An electronic device comprising a memory and a processor, characterized in that The memory stores a computer program, and the processor is configured to execute the computer program to execute the method of any one of claims 1-2.

Images